otel-audit

Scan repositories for OpenTelemetry instrumentation and generate coverage reports.

12|8|Updated Mar 20, 2026
One-click install
npx skills add https://github.com/signalfx/obstudio --skill otel-audit
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: otel-audit
Source: https://github.com/signalfx/obstudio/tree/main/skills/otel-audit
Command: npx skills add https://github.com/signalfx/obstudio --skill otel-audit

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It helps you quickly identify what OpenTelemetry instrumentation already exists in a codebase and exactly what’s missing, so you can avoid guesswork before making changes.

Core Features & Use Cases

  • Repository-wide instrumentation discovery: Detects the language/framework, enumerates HTTP routes, and records concrete evidence paths (manifests, entry points, route sources, and runtime/startup files).
  • Signal-level audit for spans, metrics, and logs: Produces explicit inventories of span sources (including individual span names), metric names, and log integration details (e.g., trace-context injection).
  • RED (Rate, Errors, Duration) gap analysis with anti-pattern detection: Classifies whether each RED dimension is covered, partial, or missing and flags common pitfalls like missing context propagation and SDK misconfiguration.
  • Generates a ready-to-read report: Writes the results to .observe/otel.md in the scanned service root, plus a concise chat summary that points to the report.

Quick Start

Run otel-audit from your service folder to create .observe/otel.md with a read-only observability coverage scan and recommendations for what to instrument next.

Frequently Asked Questions about otel-audit

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I find OpenTelemetry instrumentation gaps in my codebase?

To find OpenTelemetry instrumentation gaps, scan the repository to detect existing spans, metrics, and logs, then generate a coverage report highlighting missing RED signals and context propagation issues.

What is an OpenTelemetry coverage audit and when do I need one?

An OpenTelemetry coverage audit inventories existing spans, metrics, and logs to identify missing observability data. You need one before adding auto-instrumentation or assessing service readiness to avoid redundant changes.

How do I audit RED signals for missing metrics and spans?

Auditing RED signals requires classifying Rate, Errors, and Duration coverage as complete, partial, or missing. The audit flags missing context propagation and SDK misconfigurations to produce actionable next steps.

Can I use an instrumentation audit to list HTTP routes before adding OpenTelemetry?

Yes, an instrumentation audit explicitly enumerates HTTP routes and records concrete evidence paths like manifests and entry points. This validation planning prevents guesswork before implementing OpenTelemetry changes.

What's the best way to troubleshoot missing OpenTelemetry traces and metrics?

Troubleshoot missing OpenTelemetry traces by running a signal-level audit that extracts line-level attribution for span names and metric sources, detecting anti-patterns like missing trace-context injection in logs.

Does an OpenTelemetry coverage scan modify my service code?

No, the OpenTelemetry coverage scan is read-only. It analyzes the service folder to write an observability report to .observe/otel.md and provides a concise chat summary without altering source files.